Linked Data Approach to Water Resources Management of Hydropower Reservoirs

Abstract

The exploitation of hydropower reservoirs involves daily water resources management. Water resources managers consult periodically collected detailed data about the condition of the hydropower reservoirs, water economic data, meteorological data and forecasts, geographical and geospatial information about the water reservoirs and their contingent environment. These data are usually scattered and looked at one by one per category. Moreover, satellite data have not been typically involved in this process. We propose a linked data approach [1] to create knowledge value chain for the needs of hydropower reservoirs exploitation, based on semantic integration of earth observation data with spatial information of GIS, symbolic, numerical data and domain knowledge. We demonstrate on the example of water balance calculation for dams and cascades how satellite data and in-situ measurements are being mixed with other data, apply neural networks [2] to forecast water levels, water volumes and other characteristics based on historic data. We show how the results of the forecasts interoperate with the linked data infrastructure, how dangerous situations for the contingent areas of the water reservoirs can be identified and how signals about risks of dam overflow and hence about the water quantities available for discharge can be provided.